This subproject is one of many research subprojects utilizing the resources provided by a Center grant funded by NIH/NCRR. The subproject and investigator (PI) may have received primary funding from another NIH source, and thus could be represented in other CRISP entries. The institution listed is for the Center, which is not necessarily the institution for the investigator. This project involves research and development of a decision support system for stroke. This goal, being a very ambitious one, is broken down into several sub-projects. The first is described below. The aim of this project is to evaluate DWI-PWI mismatch hypothesis. The approach is to differentiate the focal, penumbral and benign oligaemic regions in case of an ischemic stroke with T2, diffusion, perfusion and T2* weighted images using artifical neural network and atlas based image segmentation algorithms. The final lesion size in the T2 weighted image (with/without reperfusion therapy) is used to evaluate the hypothesis. Whether DWI accurately identifies permanently damaged tissue will be established using the final lesion size of the T2 image in case where reperfusion is not administered. The accuracy of the DWI-PWI mismatch in indicating the penumbral region will be tested in case where reperfusion therapy is administered to the patient.